1. Field of the Invention
The present invention relates to a disc, and, more particularly, to an optical recording medium, an apparatus and a method of recording/reproducing data thereon/therefrom, and a computer-readable recording medium storing a program to perform the method.
2. Description of the Related Art
Recording data on, or reproducing data from, an information storage medium such as a hard disc (HD), a compact disc (CD), or a digital versatile disc (DVD) is performed in regular units. The units are recording unit blocks or reproducing unit blocks. An example of the recording unit block or the reproducing unit block is an error correction block to correct an error generated when recording or reproducing data.
When only 4 Kbytes of a 64 Kbytes error correction block are recorded, the rest of the error correction block, that is, the remaining 60 Kbytes, is filled with invalid values such as 00h.
When a disc drive reads the error correction block with only a portion of the block having valid data, and then performs error correction to reproduce the error correction block, the entire error correction block is determined to be uncorrectable because code words including the 60 Kbytes of invalid data cannot be corrected, even though code words including the 4 Kbytes of valid data can be corrected. The disc drive cannot distinguish which part of the error correction block is filled with valid data and which part of the error correction block is filled with invalid data. Therefore, the disc drive performs error correction or reproduction again, or erroneously finalizes the error correction block, even if a part of the error correction block has valid data. This situation occurs not only when reproducing data but also when updating data or adding data.
U.S. Pat. No. 6,367,049 (hereinafter referred to as “interleaving encoding”) discloses a method of encoding data using an interleaving method. The “interleaving encoding” interleavingly arranges an “LDS block,” including user data, and a “BIS block,” including address data, in a physical cluster and then records them. When reproducing data, the “BIS block” is error corrected first, and then the “LDS block” is error corrected. Referring to FIG. 1, the “interleaving encoding” will be briefly described.
FIG. 1 is a schematic representation of a conventional method of interleaving encoding.
Referring to FIG. 1, user data 11 received from a source (not shown) such as a host or an application is divided into data frames each having 2048+4 bytes. The user data 11 forms a data block 12 arranged in 304 columns and 216 rows. Next, a long distance (LDS) block 13 is formed by adding 32 parity rows to the data block 12. Then, the LDS block 13 is arranged in 152 columns and 496 rows, forming an error correcting code (ECC) cluster 14. The ECC cluster 14 is dispersed to fill sections of a physical cluster block 20 labeled ECC.
Logical address and control data 15 combined by a recording system is arranged in 32*18 bytes. A physical address 16, related to a physical location of data on a medium, is arranged in 16*9 bytes. The logical address and control data 15 and the physical address 16 are combined to form an access block 17 with 24 columns*30 rows. Next, 32 parity rows are added to the access block 17 to form a burst indicator subcode (BIS) block 18. The BIS block 18 is arranged in a BIS cluster 19 having 3 columns and 496 rows. The BIS cluster 19 is dispersed to fill sections of the physical cluster block 20 labeled BIS. A synchronizing bit group of one column is added to the physical cluster block 20, thereby forming 155 columns*496 rows of the physical cluster block 20. By arranging data by interleaving, error correction ability is improved.
When recording data on a disc that is a recordable information storage medium, a drive system records data in cluster units, which are recording units. When a cluster is formed in, for example, 32 sectors, the drive system pads some invalid sectors, makes the number of the sectors a multiple of the cluster to match the number of the sectors of the cluster if the size of the sector that needs to be recorded is not a multiple of 32, and then records the data.
Also, even when adding data or updating data to an already recorded cluster when recording a portion of a sector (e.g., 16 sectors), and not a unit of a cluster (i.e., 32 sectors), the drive system reads a cluster comprising 32 sectors, including the 16 sectors in which data is to be added or updated from a disc 21, and stores the cluster in an internal memory. Then, after the cluster is error corrected, the 16 sectors in which data is to be added or updated are modified to a corresponding location in the internal memory, encoded together with the rest of the 16 sectors into one ECC cluster, and written on the disc 21. This process is called “read-modify-write,” and is illustrated in FIG. 2. Of course, data will be recorded on the same physical address if the cluster does not have a defect, but data will be recorded on an alternative cluster according to a configured defect management method if the cluster has a defect when reproducing the cluster in the recordable information storage medium. In a write-once recording medium, data will be written on an alternative cluster by a defect process due to the characteristics of the write-once recording medium on which data can be written only once.
An operation of adding data to or updating data in 16 sectors to an already recorded cluster in a data structure, as in the “interleaving encoding” method, will be described in more detail. When reading a disc in a read-modify-write process to add data to or update data in the 16 sectors, data of a recording/reproducing cluster unit is read from the disc and stored in an internal memory. Then, after error correcting a BIS cluster first, an LDS cluster for data of 32 sectors is error corrected. If the error correction of the LDS cluster is unsuccessful, the rest of the 16 recorded sectors besides the 16 sectors in which data is to be added or updated cannot be reproduced, and thus data that is to be added or updated cannot be recorded. This is because a complete error correction needs to be performed, since valid data may be in the rest of the 16 sectors besides the 16 sectors in which data is to be added or updated. Therefore, in this case, if the validity of the data of the rest of the 16 sectors is not known, the drive system accesses the cluster in the disc again and retries reproduction, or transmits an error message to a host if the cluster cannot be reproduced despite the retry. Consequently, in such situations, the drive system is needlessly operated.
In the above situation, if the drive system knows the validity of the data of the rest of the 16 sectors besides the 16 sectors in which data is to be added or updated, and that the data is invalid, the drive system can add new data to or update the data in the 16 sectors in which data is to be added or updated without having the drive system unnecessarily retry reproduction or transmit an error message to the host. However, the validity of the rest of the 16 sectors cannot be known, and thus there is a high possibility for the drive system to error process the rest of the 16 sectors. As a result, error-correcting capability decreases.
Meanwhile, the drive system performs disc certification to examine whether the disc has a defect. The disc certification is to examine whether there is a defect in a cluster by the drive system padding an entire cluster by a predetermined value (00h or FFh) known to the drive system, recording a location of a cluster that is to be certified, and then reproducing the cluster. This is to learn a defect status of an entire disc at the initialization of the disc to use the disc, to re-learn a defect status of the entire disc at a re-initialization of the disc to use the disc from the start while using the disc, or to learn a defect status of a particular area while using the disc when needed. In addition, the drive system updates defect information of the cluster by reflecting the results of the certification.
FIG. 3 illustrates a conventional method of disc certification.
Referring to FIG. 3, when a disc 30 comprising a lead-in area 31, a data area 32, and a lead-out area 33 is certified, the entire data area 32 of the disc 30 is filled with, for example, “0.”
After certifying the disc 30 to detect a defect status of a portion of the disc 30 or the entire disc 30, the drive system knows that an area that is certified is an already written area by an RF signal when trying to write data in the area that is certified, since at least the area that is certified is an already written area. Therefore, when adding or updating 16 sectors, which are a part of a cluster comprising 32 sectors, to an already written cluster, a read-modify-write process is needed. In the read-modify-write process, when reading the disc 30, the drive system reads data in the recording/reproducing unit cluster and stores the data in the internal memory, and after error correcting the BIS cluster first, the LDS cluster is error corrected. If the LDS cluster cannot be error corrected, the rest of the 16 sectors apart from the 16 sectors to be added or updated cannot be reproduced, and so the data that is to be added or updated cannot be written. Therefore, the drive system re-accesses the cluster in the disc 30 and retries reproduction, or transmits an error message to the host if the cluster cannot be reproduced even after the second try.